Zyte MCPAll of Zyte,inside your agent

Search, fetch and extract the pages your agent can't.Build, run and monitor your crawls, all from the agent you already work in.

Run this in your terminal to add Zyte as a remote MCP server in Claude Code.

$ claude mcp add --transport http zyte https://mcp.zyte.com/v1/mcp

Then, in Claude Code, run /mcp and sign in to Zyte.

Capabilities

What your agent can do

Tools your agent can call through Zyte MCP.

Get the pages your agent can't reach on its own, back as clean Markdown it can read straight away, as rendered HTML or as a screenshot

  • Clean Markdown of the main content, with headings, lists, tables and links kept
  • Any method, headers and body over plain HTTP
  • A browser that can click, type, scroll and run JavaScript first
  • Choose the location, keep cookies and sessions

Search the web and return results as structured data

  • Search as seen from a given country or language
  • Describe the fields you want in plain language, on any page
  • Works on raw or browser-rendered pages

Check the price of scraping a site with Zyte API

  • What a request to it costs

Start spiders and scripts on Scrapy Cloud, and manage the schedules that run them

  • Start a spider or a script
  • Create, pause, change and delete schedules
  • Find, cancel and tag jobs

Read a Scrapy Cloud job's state, counters, log, requests and items without leaving the conversation

  • Job state and counters
  • The log, requests and items of a job
  • How many items have each field

Your Zyte API spend, success rate and response times, as a summary or broken down by date, domain or feature

  • Spend and request volume
  • Success rate and response times
  • By date, domain or feature, over a window you choose

Walkthrough

See Zyte MCP in action

FAQ

Questions about Zyte MCP

Zyte MCP itself is free. What your agent does through it is billed to your Zyte account at Zyte API prices, and Scrapy Cloud jobs it starts are billed at Scrapy Cloud prices. One request to your agent can turn into many Zyte API requests, for example one per page of a paginated listing. New users get up to 30,000 requests free.

Small print: $5 free credit Zyte API trial.

Your agent can act on any Scrapy Cloud project your account can access. That includes starting and cancelling jobs, changing schedules and settings, and writing and deleting collection records. Deleting a schedule, a collection or its records can't be undone. Your agent is told to confirm with you before any change, so review what it asks to do before you approve it.

When you sign in, your agent's Zyte API requests use their own key, mcp_access_key. Filter your Zyte API stats by that key to see what the agent spent, and set a spending limit on it to cap that spending. Jobs your agent starts on Scrapy Cloud aren't covered by that limit.

Agents have their own tools for fetching pages and searching, and they often reach for those first. Naming Zyte in your request, as in "fetch this URL with Zyte", is usually enough. For pages that need a browser or a screenshot, or for data extraction, agents tend to choose Zyte on their own.